Pericentrin deficiency in smooth muscle cells augments atherosclerosis through HSF1-driven cholesterol biosynthesis and PERK activation

Abstract

Microcephalic osteodysplastic primordial dwarfism type II (MOPDII) is caused by biallelic loss-of-function variants in pericentrin (PCNT), and premature coronary artery disease (CAD) is a complication of the syndrome. Histopathology of coronary arteries from patients with MOPDII who died of CAD in their 20s showed extensive atherosclerosis. Hyperlipidemic mice with smooth muscle cell-specific (SMC-specific) Pcnt deficiency (PcntSMC-/-) exhibited significantly greater atherosclerotic plaque burden compared with similarly treated littermate controls despite similar serum lipid levels. Loss of PCNT in SMCs induced activation of heat shock factor 1 (HSF1) and consequently upregulated the expression and activity of HMG-CoA reductase (HMGCR), the rate-limiting enzyme in cholesterol biosynthesis. The increased cholesterol biosynthesis in PcntSMC-/- SMCs augmented PERK signaling and phenotypic modulation compared with control SMCs. Treatment with the HMGCR inhibitor, pravastatin, blocked the augmented SMC modulation and reduced plaque burden in hyperlipidemic PcntSMC-/- mice to that of control mice. These data support the notion that Pcnt deficiency activates cellular stress to increase SMC modulation and plaque burden, and targeting this pathway with statins in patients with MOPDII has the potential to reduce CAD in these individuals. The molecular mechanism uncovered further emphasizes SMC cytosolic stress and HSF1 activation as a pathway driving atherosclerotic plaque formation independently of cholesterol levels.

Keywords: Atherosclerosis; Cell stress; Cholesterol; Genetics; Vascular Biology.

Figure 1. Characterization of coronary artery atherosclerotic lesions in patients with MOPDII.

(A) Left anterior descending coronary artery from MOPDII Patient 1. Low-magnification image of Movat staining (a) reveals a large atherosclerotic plaque (P) in the left anterior descending coronary artery with a small lumen (L). Higher magnification image of the medial layer (M), shown in b and c, shows a well-formed internal elastic lamina (c, blue arrow), with an absence of cell staining in the medial layer and increased elastin deposition in the adventitial layer (Ad; the black elastin fibers are found in the adventitial layer). H&E staining (d–f) confirms loss of medial cells, along with cholesterol crystals in the atherosclerotic plaque. α-Smooth muscle actin (SMA) staining reveals large numbers of SMA-positive cells in the adventitia and fibrous cap (g and h) of the atherosclerotic plaque and an absence of SMA-positive cells in the medial layer (i). (B) Left main coronary artery bifurcation of MOPDII Patient 2. The low-magnification image of the Movat staining (a) reveals a large atherosclerotic plaque (P) with residual lumen (L). Higher-magnification images of the artery wall reveal an absence of cells in the medial layer (M) and elastosis (b and c). H&E staining (d–f) shows similar changes. SMA staining shows SMA-positive cells in a portion of the artery wall (g) and in the fibrous cap (h), with a paucity of SMA-positive cells in other areas of the medial wall (i). Scale bars: 500 μm (left), 200 μm (middle), and 50 μm (right).

Figure 2. SMC-specific Pcnt-deficient mice have increased atherosclerotic plaque burden.

(A and B) En face Oil Red O staining of aortas shows significantly increased plaque formation in hyperlipidemic Pcnt^SMC–/– mice placed on 12 weeks of HFD, compared with similarly treated WT mice (n = 10–12 mice per genotype per sex, Mann-Whitney U test). (C) Only male Pcnt^SMC–/– mice had significantly higher plaque burden compared with male WT mice, and female mice showed no statistically significant difference (n = 10 males and 12 females, 2-way ANOVA followed by Tukey’s multiple-comparison test). (D–F) H&E staining (D) demonstrates that male Pcnt^SMC−/− mice had greater atherosclerotic lesion areas in both the aortic roots (E) and ascending aortas (F) (n = 6, by unpaired, 2-tailed Student’s t test with Welch’s correction). Scale bars: 200 μm (bottom left) and 500 μm (all others). (G–I) SMA staining of the aorta (G) reveals significantly fewer SMA⁺ cells in the medial layer (H) but not the plaque (I) of the aortic root (n = 9, Mann-Whitney U test). Error bars represent SD. *P < 0.05, **P < 0.01, ****P < 0.0001. L, lumen; M, medial layer; P, plaque. Scale bars: 5 μm.

Figure 3. Pcnt^SMC–/– mice show significantly increased expression of SMC modulation markers in aortic root lesions.

(A–E) Immunohistochemical staining of aortic root sections for SMC modulation markers LGALS3, FN1, PAI1, VCAM1, and SCA1 and the differentiation marker SMA shows significantly higher staining for modulation markers and lower staining for SMA in Pcnt^SMC–/– mice compared with WT mice. Nuclei were counterstained with DAPI (blue). (F) Immunohistochemical staining of aortic root sections for the macrophage marker F4/80 shows no change. (G–K) Quantification of the results in A–E (n = 9, quantification for LGALS3, FN1, PAI1, and SCA1 was analyzed by unpaired, 2-tailed Student’s t test followed by Welch’s correction and that for VCAM1 was analyzed by Mann-Whitney U test). (L) Quantification of the results in F (n = 9, unpaired, 2-tailed Student’s t test followed by Welch’s correction). Error bars represent SD. ****P < 0.0001. L, lumen; M, medial layer; P, plaque. Scale bar: 20 μm.

Figure 4. Augmented phenotypic modulation of Pcnt^SMC–/– SMCs is due to increased HSF1 activation driving cholesterol biosynthesis.

(A) Augmented SMC phenotypic modulation in Pcnt^SMC–/– SMCs is evident from decreased mRNA expression of Cnn1 and increased mRNA expression of modulation markers Lgals3, Fn1, Serpine1, and Ly6a either at baseline or with exposure to 2.5 μg/mL MBD-Chol in Pcnt^SMC–/– SMCs, compared with 10 μg/mL MBD-Chol in WT SMCs, while Vcam1 expression increases in both genotypes only with 10 μg/mL MBD-Chol. (B) Pcnt^SMC–/– SMCs show increased migration by Transwell migration assay at baseline and with cholesterol exposure. (C) Pcnt^SMC–/– SMCs exhibit increased levels of total and phosphorylated HSF1 (p-HSF1) at baseline. (D) HSF1 mRNA expression and luciferase activity and expression of HSF1 downstream targets Hspa1a, Hsp90aa1, and Hsp90ab1 are also upregulated at baseline in Pcnt^SMC–/– SMCs (all data in A–D passed normality and were analyzed by 2-way ANOVA followed by Tukey’s multiple-comparison test, except HSF1 activity, which was analyzed by Kruskal-Wallis test). (E–G) Hmgcr expression was significantly elevated in Pcnt^SMC–/– SMCs (E), along with HMGCR enzymatic activity (F) and cholesteryl ester levels (G; data in E–G were analyzed by unpaired, 2-tailed Student’s t test followed by Welch’s correction). All gene expression data are representative of 3 independent experiments. Error bars represent SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. CE, cholesteryl esters.

Figure 5. Pcnt deletion–induced augmented PERK signaling and SMC phenotypic modulation are HSF1 dependent.

(A and B) Pcnt^SMC–/– SMCs have increased Atf4 expression and KLF4 luciferase activity (A), and increased levels of eIF2α phosphorylation, ATF4, and KLF4 (B), at baseline or with exposure to 2.5 μg/mL MBD-Chol in Pcnt^SMC–/– SMCs, compared with 10 μg/mL MBD-Chol in WT SMCs. (C) HSF1 activity is significantly decreased in both WT and Pcnt^SMC–/– SMCs following siRNA-mediated depletion of Hsf1. (D) Hmgcr expression, HMGCR enzymatic activity, and cholesteryl ester levels are significantly reduced in Pcnt^SMC–/– SMCs following siRNA-mediated depletion of Hsf1. (E–G) siRNA-mediated depletion of Hsf1 reduces activation of the PERK pathway (E), and phenotypic modulation (F and G) at baseline. All gene expression data are representative of 3 independent experiments. Multiple group comparisons were analyzed by 2-way ANOVA followed by Tukey’s multiple-comparison test. Error bars represent SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. CE, cholesteryl esters.

Figure 6. Treatment with the HMGCR inhibitor pravastatin reverses augmented SMC modulation and reduces the plaque burden in Pcnt^SMC–/– mice.

(A) Treatment with the HMGCR inhibitor pravastatin significantly reduces HMGCR activity and cholesteryl ester levels. (B–D) Pravastatin treatment suppresses baseline activation of the PERK pathway (B) and phenotypic modulation in Pcnt^SMC–/– SMCs (C) and also decreases migration in both Pcnt^SMC–/– and WT SMCs (D). (E and F) Treatment with pravastatin reduces plaque burden to similar levels in Pcnt^SMC–/– and WT mice (n = 10 males). All gene expression data are representative of 3 independent experiments. Multiple group comparisons for both cellular and animal data were analyzed by 2-way ANOVA followed by Tukey’s multiple-comparison test. Error bars represent SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. CE, cholesteryl esters.